Digital filter and method for performing a multiplication based on a look-up table

A method for performing digital filtering of input sample data, comprising the steps of:

• a) delaying said input sample data by predetermined delay periods to generate delayed input sample data;
• b) pre-storing weighted sample data corresponding multiplications of input sample data with predetermined filter coefficients;
• c) generating other weighted input data corresponding to other multiplications of input sample data with predetermined filter coefficients by performing a bit shifting operation using said pre-stored weighted sample data, wherein the values of said other weighted sample data correspond to the values of said pre-stored weighted sample data multiplied by two or a power of two,
• wherein said other weighted sample data correspond to an input sample data, the value of which can be obtained by multiplying another input sample data by a factor equal to two or a power of two, and are generated by using a corresponding one of said pre-stored weighted sample data relating to said other input sample data and performing a corresponding bit shifting operation in accordance with said multiplication factor, and
• further comprising the step of storing said prestored weighted sample data in look-up tables each corresponding to one filter coefficient, and providing an index table addressing said other input sample data, and a zero pad table indicating the number of bit shifting operations to be performed on said corresponding one of said pre-stored weighted sample data relating to said other input sample data.

A method for performing a multiplication based on a look-up table, comprising the steps of:

• a) obtaining those multiplication results, which correspond to a multiplier or multiplicand value which can be obtained by a multiplying another multiplier or multiplicand value by a factor equal to two or a power of two, by using the multiplication result corresponding to said another multiplier or multiplicand value and performing a corresponding a bit shifting operation in accordance with said factor;
• and
• b) storing only those multiplication results which cannot be generated from another multiplication result by said bit shifting operation,
• further comprising the step of storing said multiplication results in look-up tables each corresponding to a single predetermined multiplier, and providing an index table addressing said another multiplicand, and a zero pad table indicating the number of bit shifting operations to be performed on said multiplication results corresponding to said another multiplicand.